TWI650478B - Method for detecting lubricating oil consumption rate of transmission mechanical components - Google Patents

Abstract

The invention provides a method for detecting a lubricating oil consumption rate of a transmission mechanical component, by using a first prediction model to predict a total operating physical quantity corresponding to an operating speed of a transmission mechanical component, and using a second prediction model to predict a corresponding transmission machinery The temperature coefficient of the temperature value of the component, and predicting the total amount of lubricant consumed by the transmission mechanical component according to the physical quantity of the operation corresponding to the actuation of the transmission mechanical component, the predicted total physical quantity of the actuation, and the predicted temperature coefficient. The number of actuations, and the number of actuations of the transmission mechanical components during a time period is calculated, the lubricant consumption rate of the transmission mechanical components during the time period can be calculated, and the time at which the transmission mechanical components need to be refilled with the lubricant can be further calculated. .

Description

Method for detecting lubricating oil consumption rate of transmission mechanical components

The present invention relates to a method for detecting the rate of consumption of lubricating oil, and more particularly to a method for detecting the rate of consumption of lubricating oil for a mechanical component of a transmission.

Due to its high positioning accuracy and long service life, the ball screw has recently been widely used in machine tools that require precise positioning. Generally, the ball screw is screwed by the nut through the rolling ball and the screw, and the nut is linearly moved relative to the screw by the rolling of the ball in the nut, so the ball screw needs to be injected through the lubricating oil. To maintain lubrication to reduce the friction between the nut and the screw, thus extending the life of the ball screw.

Therefore, how to automatically detect that the lubricating oil injected into the ball screw has been consumed and re-injecting the lubricating oil is an important issue. The Republic of China Patent Publication No. I525409 discloses a method for judging the timing of oiling of a linear transmission component by judging a vibration sensor sensed on a ball screw and analyzing a vibration signal sensed by the vibration sensor to determine a lubricant of the ball screw. When the consumption is completed. However, the foregoing method needs to use the vibration sensor to sense the vibration signal and perform calculation and analysis when the ball screw is in operation, which not only increases the hardware cost but also requires more computing resources.

Accordingly, it is an object of the present invention to provide a method of detecting the rate of consumption of lubricating oil for a transmission mechanical component.

Therefore, the method for detecting the lubricating oil consumption rate of the transmission mechanical component of the present invention is implemented by a detecting system comprising a computing device disposed in a machine tool and injecting a quantity of lubricating oil, and The transmission mechanical component corresponds to an actuation speed during operation. The method for detecting the lubricating oil consumption rate of the transmission mechanical component comprises a step (a), a step (b), a step (c) and a step (d).

The step (a) is that the computing device utilizes a first predicted model of a corresponding amount of lubricating oil prepared in advance, and predicts a total operating physical quantity corresponding to the operating speed according to the operating speed, wherein the total operating physical quantity is The sum of the actuating physical quantities obtained by consuming the amount of lubricating oil to actuate the transmission mechanical component under the condition of the operating speed, and the actuating physical quantity is the actuating physical quantity obtained by actuating the transmission mechanical component once.

The step (b) is that the computing device predicts a total number of actuations corresponding to the actuation speed based on at least the actuation physical quantity and the predicted total actuation physical quantity.

The step (c) is that the computing device receives the actuation information for a corresponding period of time from the machine tool, and calculates the number of actuations of the transmission mechanical component during the time period based on the actuation information.

The step (d) is that the computing device calculates a lubricating oil consumption rate during a corresponding time period based on the number of actuations of the transmission mechanical component during the time period and the predicted total number of actuations.

The effect of the invention is that the lubricating oil consumption rate of the transmission mechanical component can be automatically estimated.

Referring to Figure 1, a first embodiment of the method for detecting the rate of lubricating oil consumption of a transmission mechanical component of the present invention is implemented by a detection system comprising a computing device 2 disposed on a machine tool 1 and injected with a quantity of lubricating oil, the computing device 2 is electrically connected to the machine tool 1, and the transmission mechanical component 11 corresponds to an operating speed during operation. For example, the transmission mechanical assembly 11 is a ball screw or a ball spline, and the actuation speed is a rotational speed.

Referring to Fig. 2, the implementation steps of the first embodiment will be described below.

First, in step S11, the computing device 2 uses a first predicted model of a corresponding amount of lubricating oil prepared in advance, and predicts a total operating physical quantity corresponding to the operating speed based on the operating speed, wherein the total operating physical quantity is The sum of the actuation physical quantities obtained by consuming the amount of lubricating oil to actuate the transmission mechanism assembly 11 under the condition of the actuation speed, and the actuation physical quantity is the actuation physical quantity obtained by the transmission mechanical assembly 11 being actuated once. .

Referring to FIG. 3, the ball screw 3 includes a screw 31 and a nut 32, and the screw 31 has a first end 311 and a second end 312. A spiral groove 33 is formed on the surface of the screw 31. When the machine tool 1 is in operation, the one action of the ball screw 3 means that the nut 32 moves from the first end 311 to the second end 312 or from the second end 312 to the first end 311. . Therefore, the actuating physical quantity obtained by one operation of the ball screw 3 is that the nut 32 moves from the first end 311 to the second end 312 or moves from the second end 312 to the first end 311. One of the surface area of the spiral groove 33, the length of the spiral groove 33, and the linear movement distance of the nut 32; and the total actuation physical quantity is that the ball screw 3 is operated by consuming the quantity of lubricating oil One of the sum of the surface area of the spiral groove obtained, the sum of the lengths of the spiral grooves, and the sum of the linear movement distances of the nuts.

Referring to FIG. 4, preferably, the first prediction model is a linear function of a first logarithmic function ln(X ₁ ), that is, Y ₁ =A×ln(X ₁ )+C ₁ , where X ₁ is the actuation Speed, Y ₁ is the total amount of physical activity corresponding to the speed of action. Before using the first prediction model, it is necessary to prepare a plurality of training materials 4 of the first prediction model to train/estimate parameters of the first prediction model, that is, A and C ₁ ; wherein the first prediction model Each training material contains two elements, namely the "actuation speed" and the "total physical quantity" corresponding to the speed of action. However, the first prediction model is not limited to the above, and may be other linear regression models or a neural network model.

The manner in which the training material 4 of the first predictive model is prepared is not unique. With the ball screw 3, the total amount of time required for the ball screw 3 to consume the quantity of lubricating oil under a specific rotational speed condition can be estimated by the method disclosed in the Republic of China Patent Publication No. I525409; Under this particular rotational speed condition, the amount of actuation physical force obtained by the rotation/movement of the ball screw 3 per unit time is fixed, so that the total operational physical quantity corresponding to the specific rotational speed can be further calculated. As shown in FIG. 4, the training materials of the first prediction model may cover a plurality of different actuation speeds and corresponding total actuation physical quantities; thus, the first prediction model may be utilized to predict the total corresponding to any actuation speed. Actuate physical quantities.

Next, in step S12, the computing device 2 predicts a total number of actuations corresponding to the actuation speed based on the actuation physical quantity and the predicted total actuation physical quantity. Preferably, the predicted total number of actuations = the predicted total actuation physical quantity ÷ the actuation physical quantity.

Next, in step S13, the computing device 2 receives an actuation information corresponding to a time period from the machine tool 1, and calculates the number of actuations of the transmission mechanism assembly 11 during the time period based on the actuation information. For example, referring to FIG. 5, the machine tool 1 drives the transmission mechanical component 11 to output a first voltage value α greater than zero when the transmission mechanical component 11 is actuated, and the voltage value output by the power tool 1 is zero when the transmission mechanical component 11 is not actuated; The computing device 2 receives from the machine tool 1 a voltage signal 6 during the corresponding time period and calculates the number 61 of peaks of the voltage signal 6 as the number of actuations of the transmission mechanical component 11 during the time period. In the example shown in Figure 5, the transmission mechanism assembly 11 is actuated 10 times in 60 seconds.

Next, in step S14, the computing device 2 calculates a lubricating oil consumption rate during a corresponding time period according to the number of actuations of the transmission mechanical component 11 during the time period and the predicted total number of actuations; wherein, the lubrication Oil consumption rate = the number of total actuations predicted by the number of actuations of the transmission mechanical component during this time period x 100%.

In particular, by using the above-described method of detecting the lubricating oil consumption rate of the transmission mechanical component, it can be further judged when the transmission mechanical component 11 has consumed the amount of lubricating oil and needs to be injected again. In detail, referring to FIG. 6, the computing device 2 receives actuation information from the machine tool 1 periodically, for example every 60 seconds, and determines, during each time period, the actuation information according to the corresponding time period. The number of actuations of the transmission mechanism assembly 11 during the period; and for each time period, the calculation device 2 calculates the lubricant consumption rate during the corresponding time period, and accumulates the lubricant consumption rates in time series to obtain an accumulation The lubricating oil consumption rate, and determining whether the accumulated lubricating oil consumption rate is greater than a predetermined upper limit value (step S15); preferably, the upper limit value is 100%. When it is judged that the accumulated lubricating oil consumption rate is greater than the upper limit value, the calculating means 2 determines that the transmission mechanical component 11 needs to be injected with lubricating oil again (step S16).

It should be particularly noted that, although the method of detecting the lubricating oil consumption rate and the time of the oiling point of the transmission mechanical component described in the first embodiment is used on-line, the training data needs to be prepared in advance to train the first. Predicting the model, but the training of the first predictive model only needs to be performed once and is off-line, and the machine tool 1 / transmission mechanical component 11 is directly used to use the first prediction that has been trained. The model, so the amount of calculation required for the first embodiment is relatively small when executed on the line. In addition, compared with the Republic of China Patent Publication No. I525409, the first embodiment does not need to use a vibration sensor to sense the vibration signal of the transmission mechanical component 11 when executed on the line, so the hardware cost and the computing resource requirement. Both are lower.

A second embodiment of the method of detecting the lubricating oil consumption rate of the transmission mechanical component of the present invention will now be described. The main difference between the second embodiment and the first embodiment is that in the second embodiment, in addition to considering the total actuation physical quantity corresponding to the quantity of lubricating oil, in addition to considering the operating speed of the transmission mechanical component 11 In addition, the influence of the temperature at which the transmission mechanical component 11 is actuated is also considered. In general, the higher the temperature value, the faster the lubricant is consumed, and the total amount of total physical activity corresponding to the amount of lubricant is also smaller.

In detail, in the second embodiment, the detection system further includes a temperature sensor disposed on the transmission mechanical component 11. In the case of the ball screw 3, as shown in FIG. 7, the temperature sensor 7 is disposed on the nut 32 of the ball screw 3.

Referring to Figures 8 and 9, the implementation steps of the second embodiment are explained below.

First, in step S21, as in step S11, the computing device 2 uses the first prediction model to predict a total actuation physical quantity corresponding to the actuation speed based on the actuation speed.

Next, in step S22, the computing device 2 receives a temperature value sensed by the temperature sensor 7 from the temperature sensor 7 for a period of time, and utilizes a predetermined amount of lubricating oil prepared in advance. The second prediction model of the actuation speed predicts a temperature coefficient corresponding to the temperature value based on the sensed temperature value. Wherein, the temperature coefficient is a ratio of the total operating physical quantity corresponding to the operation of the transmission mechanical component 11 under the condition of the temperature value, and the total operating physical quantity corresponding to the operation under a reference temperature condition, and the reference temperature Not greater than this temperature value. For example, suppose the reference temperature and the temperature value are respectively 20 ° C, 30 ° C, and the transmission mechanical component 11 is operated at a temperature of 3000 ° C under the condition of a temperature value of 20 ° C and 30 ° C. The operating physical quantities are M and N, respectively, and for a rotational speed of 3000 RPM, the temperature coefficient corresponding to the temperature value of 30 ° C is N/M.

As shown in FIG. 9, the second prediction model is a linear function of a second logarithmic function ln(X ₂ ), that is, Y ₂ =B×ln(X ₂ )+C ₂ , where X ₂ is sensed. The temperature value to which Y ₂ is the temperature coefficient corresponding to the temperature value. Before using the second prediction model, for the actuation speed, a plurality of training materials 8 of the second prediction model need to be prepared to train/estimate the parameters of the second prediction model, that is, B and C ₂ ; Each training material of the second prediction model contains two elements, namely the "temperature value" and the "temperature coefficient" corresponding to the temperature value. In particular, the training data sets corresponding to different actuation speeds are different, so the parameters of the second prediction model trained are different. However, the second prediction model is not limited to the above, and may be other linear regression models or a neural network-like model.

The manner in which the training material 8 of the second prediction model is prepared is not unique. In the case of the ball screw 3, the temperature of the ball screw 3 can be measured in combination with the method disclosed in the Republic of China Patent Publication No. I525409, and recorded at different temperature values, the ball screw 3 is at the operating speed. The total operating physical quantity obtained by consuming the quantity of lubricating oil under the condition; then, the temperature coefficient corresponding to the different temperature values under the condition of the operating speed is further calculated.

Next, in step S23, the computing device 2 predicts the total number of actuations based on the predicted temperature coefficient, the actuation physical quantity, and the predicted total actuation physical quantity; wherein, the predicted total number of actuations = predicted The total amount of physical activity × the predicted temperature coefficient ÷ the physical quantity of the action.

Then, step S24 and step S25 are performed; the two steps are the same as steps S13 and S14 of the first embodiment, respectively, and therefore are not described herein.

In addition, as in the first embodiment, the computing device 2 periodically receives the actuation information from the machine tool 1, and accumulates the lubricating oil consumption rate over a plurality of time periods in time series to obtain an accumulated lubricating oil consumption rate, and It is judged whether or not the accumulated lubricating oil consumption rate is greater than the predetermined upper limit value, and the second embodiment can be used to judge the time point at which the transmission mechanical component 11 has consumed the same amount of lubricating oil and needs to be injected again.

In addition, it should be particularly noted that although the second embodiment needs to additionally provide the temperature sensor 7 compared to the first embodiment, since the temperature influence of the transmission mechanical component 11 is considered, The estimate of the lubricant consumption rate is more accurate. Compared with the Republic of China Patent Publication No. I525409, the hardware cost of the temperature sensor 7 is also lower than the hardware cost of the vibration sensor; in addition, the calculation amount required for the second embodiment is also low.

In summary, the present invention detects a lubricating oil consumption rate of a transmission mechanical component by using the first prediction model to predict the total operating physical quantity corresponding to the operating speed of the transmission mechanical component, and using the second prediction The model predicts the temperature coefficient corresponding to the temperature value of the transmission mechanical component, and predicts the total number of actuations according to the actuation physical quantity, the predicted total actuation physical quantity, and the predicted temperature coefficient, and according to the machine tool The actuation information calculates the number of actuations of the transmission mechanical component during the time period, calculates the lubricating oil consumption rate of the transmission mechanical component during the time period, and further calculates that the transmission mechanical component needs to be refilled with the lubricating oil. At the time, it is indeed possible to achieve the object of the present invention.

However, the above is only the embodiment of the present invention, and the scope of the invention is not limited thereto, and all the equivalent equivalent changes and modifications according to the scope of the patent application and the patent specification of the present invention are still The scope of the invention is covered.

1‧‧‧Tooling machine

11‧‧‧ Transmission mechanical components

2‧‧‧ Computing device

3‧‧‧Ball screw

31‧‧‧ screw

311‧‧‧ first end

312‧‧‧ second end

32‧‧‧ nuts

33‧‧‧Spiral groove

4‧‧‧ Training materials for the first predictive model

6‧‧‧ voltage signal

61‧‧‧Crest

7‧‧‧Temperature Sensor

8‧‧‧ Training materials for the second prediction model

Α‧‧‧first voltage value

S11~S16‧‧‧Steps

S21~S25‧‧‧Steps

Other features and advantages of the present invention will be apparent from the embodiments of the drawings, wherein: Figure 1 is a block diagram showing a transmission mechanical assembly disposed on a machine tool; Figure 2 is a flow chart illustrating A first embodiment of the method for detecting the lubricating oil consumption rate of a transmission mechanical component of the present invention; FIG. 3 is a schematic view showing a ball screw; FIG. 4 is a schematic diagram illustrating a first prediction model and corresponding plurality of training materials Figure 5 is a schematic diagram showing a voltage signal outputted by the machine tool; Figure 6 is a flow chart illustrating the use of the first embodiment to determine the point at which the transmission mechanical component needs to be refilled with lubricating oil; Figure 7 is a Schematic illustration of a temperature sensor disposed on the nut of the ball screw; FIG. 8 is a flow chart illustrating a second embodiment of the method of detecting the rate of lubricating oil consumption of the transmission mechanical assembly of the present invention; and FIG. A schematic diagram illustrating a second prediction model and corresponding plurality of training materials.

Claims (7)

A method for detecting a lubricating oil consumption rate of a transmission mechanical component is implemented by a detection system including a computing device disposed in a machine tool and injecting a quantity of lubricating oil, and the transmission The mechanical component corresponds to an actuation speed during operation, and the method for detecting the lubricating oil consumption rate of the transmission mechanical component comprises the following steps: (a) the computing device utilizes a first predicted model of a corresponding amount of lubricating oil prepared in advance, according to The actuation speed predicts a total actuation physical quantity corresponding to the actuation speed, wherein the total actuation physical quantity is an actuation physical quantity obtained by consuming the quantity of lubricating oil to actuate the transmission mechanical component under the condition of the actuation speed a summation, and the actuating physical quantity is an actuating physical quantity obtained by actuating the transmission mechanical component once; (b) the computing device predicts a total corresponding to the actuating speed based on the actuating physical quantity and the predicted total actuating physical quantity Number of actuations; (c) when the computing device receives a corresponding one from the machine tool Actuation information during the period, and calculating, based on the actuation information, the number of actuations of the transmission mechanical component during the time period; and (d) the computing device is based on the number of actuations of the transmission mechanical component during the time period and predicted The total number of actuations calculates a lubricant consumption rate for a corresponding period of time. The method for detecting a lubricating oil consumption rate of a transmission mechanical component according to claim 1, wherein the detecting system further comprises a temperature sensor disposed in the transmission mechanical component, wherein in the step (b), the computing device is Receiving, from the temperature sensor, a temperature value sensed by the temperature sensor during the time period, and using a pre-prepared corresponding amount of lubricating oil and a second prediction model of the actuation speed, according to the temperature value Determining a temperature coefficient corresponding to the temperature value; and the computing device predicts the total number of actuations based on the actuation physical quantity, the predicted total actuation physical quantity, and the predicted temperature coefficient, wherein the total number of actuations is positively correlated The predicted temperature coefficient. The method for detecting a lubricating oil consumption rate of a transmission mechanical component according to claim 2, wherein in step (b), the temperature coefficient is a total physical quantity corresponding to the operation of the transmission mechanical component under the condition of the temperature value, The ratio of the total actuation physical quantity corresponding to operation at a reference temperature, wherein the reference temperature is not greater than the temperature value. The method for detecting a lubricating oil consumption rate of a transmission mechanical component according to claim 2, wherein in the step (a), the first prediction model Y1=A×ln(X1)+C1 is a first logarithmic function ln ( X1) a linear function, wherein the actuation velocity X1 for the input of a first logarithmic function, a with C ₁ for the first parameter of the prediction model; and the step (B), the second predictive model Y ₂ = B ×ln(X ₂ )+C ₂ is a linear function of a second logarithmic function ln(X ₂ ), wherein the temperature value X ₂ is an input of the second logarithm function, and B and C _{2 are} the second prediction model parameter. The method for detecting a lubricating oil consumption rate of a transmission mechanical component according to claim 1, wherein in the step (a), the first prediction model Y1=A×ln(X1)+C1 is a first logarithmic function ln ( X1) a linear function, wherein the actuation velocity X1 for the input of a first logarithmic function, a with C ₁ for the first parameter prediction model. The method for detecting a lubricating oil consumption rate of a transmission mechanical component according to claim 1, wherein the transmission mechanical component is a ball screw including a screw and a nut, and a surface of the screw is formed with a spiral groove, and the operating speed is For the rotation speed of the ball screw, wherein in the step (a), the total actuation physical quantity is the sum of the surface area of the spiral groove experienced by the nut moving on the screw, the sum of the lengths of the spiral grooves, and the linear movement distance of the nut One of the sums. The method for detecting a lubricating oil consumption rate of a transmission mechanical component according to claim 1, wherein the computing device periodically receives the actuation information from the machine tool, and for each time period, according to the actuation information during the corresponding time period. Determining the number of actuations of the transmission mechanical component during the time period; for each time period, the computing device calculates the lubricating oil consumption rate during the corresponding time period, and accumulates the lubricating oil consumption rates according to the time sequence An accumulating lubricating oil consumption rate, and when the accumulated lubricating oil consumption rate is greater than a predetermined upper limit value, the computing device determines that the transmission mechanical component is to be injected with lubricating oil.